Joule Thief 101

[MileHigh]

Not trolling and I am not ignorant. ( well maybe somewhat ignorant )

Your best action would be to state the obvious, your question was in error and as the question was stated described an impossible condition.

The question was not in error and it's just ignorance again.

[MileHigh]

I understand that you could be essentially attempting to 'set the stage' and prime someone like myself with the 'simple basics' before moving to a real circuit, but at this point we should be able to admit it has become more complicated than necessary.. If we are to forget the R=0 business and move on, continuing to analyze a circuit that does not closely parallel anything we have going on in the real world right now,  I personally have no desire to commit because it totally derails me from the information I came into the forum looking for.

Earlier on in this thread I clearly demonstrated to you that this can be replicated on the bench with 99.99% similarity to what is being discussed here.  All that you have to do is reduce the inductance, lower the voltages and shorten the timing.  That is the real world.

And I will tell you one more time, this directly applies to building pulse motors and Joule Thieves.  If you can't see that right now, it doesn't matter.  You would have to figure that out for yourself.  Knowledge builds upon knowledge.

[Magneticitist]

Earlier on in this thread I clearly demonstrated to you that this can be replicated on the bench with 99.99% similarity to what is being discussed here.  All that you have to do is reduce the inductance, lower the voltages and shorten the timing.  That is the real world.

And I will tell you one more time, this directly applies to building pulse motors and Joule Thieves.  If you can't see that right now, it doesn't matter.  You would have to figure that out for yourself.  Knowledge builds upon knowledge.

But you did not demonstrate that.. that is the problem! this is the entire debate.

following the scientific method you have constructed your hypothesis
that it will be 99% similar in theory as it would be in the real world.

Until you can obtain an impossible inductor that can be placed within a
circuit that has absolutely 0 resistance, you have yet to even test it in
an experiment. You may have tested a very low resistance coil, so you know,
you can establish that as more than hypothesis.

The 0 resistance circuit you cannot. It shall remain hypothesis until proven
otherwise. Were you to simply concede to that infallible logical conclusion,
some of us would probably provide a little more adherence to the remainder
of your intended lesson. The issue at hand most members (from what I can tell)
at this point have with the argument, is that you are unwilling to concede to
there being no possible way to gain absolute surety about any aspects
of a circuit analysis if we include a variable of 0 resistance, and leave the
rest of the many other possible variables as vague exclusions.

All of that could have easily be sidestepped were you to simply avoid real world
comparisons and state that your test parameters include physics where R=0
and R=.00000000001 yield no great distinction deserving concern.

[partzman]

Here is another sim which will be my last on this subject unless requested otherwise.

This shows the same 5 H inductor with a dcr = 1e-15 ohms plus stepped dcr's of .1, .5, and 1 ohm respectively.  The resulting current traces are marked on the plot and as can be seen, the current reaches a lower peak level with each increase in dcr. These values can be calculated by using the time constant formula, delta I = Vin/dcr(1-1/Eu^t/tau).

Eu is Euler's constant = 2.7182818 .

t = time Vin is applied in seconds .

tau = time constant = L/R

Solving for the .5 ohm step, we have 4/.5*(1-1/2.7182818^3/10) = 8*(1-1/2.7182818^.3) = 8*(1-1/1.3498588) = 8*.2591817 = 2.07345 amps which agrees with the sim.

The observation to be made here is that as the dcr decreases, the peak current reached in the inductor under the same given conditions increases but always starts at zero. IOW, the trend as we approach zero is that we see more of a pure inductance. 

My questions then is, at what point do we all of a sudden lose the property of inductance as we decrease the dcr and become either a short circuit or an infinite inductance?

I might add that with my current LtSpice parameter set, I can run this simulation successfully with a dcr = 1e-320 with the same results as dcr = 1e-15.  With a dcr = 1e-325 an error message states it can not properly solve the matrix needed to complete the simulation.

Just for comparison, the current super-conducting electromagnets used in MRI scanners have a typical dcr in the range of 1e-11 to 1e-12.

partzman

[MileHigh]

Magneticitist:

I am not going to argue this any more.  You are just going to have to accept or cope with the fact that people can discuss circuit behaviour using ideal components and then very easily apply that knowledge to the actual circuits that they are working with on the bench.  Your argument is invalid, and I have heard it countless times before from people in a similar position to you.

The question as posed without a single modification can be answered with 100% accuracy.  Then, a keener could get a big negative feedback operational amplifier, insert a resistor for current sensing just before the output of the amplifier and the voltage sense, and then make extremely accurate measurements of the ideal voltage and associated current being put into the device under test.  For example, the device under test could be a single low-resistance coil or some kind of filtering circuit.

If you don't accept this, then there is nothing that I can do about it.  It's not for me to convince you, it's for you to convince yourself and change your attitude and embrace knowledge and embrace stuff that will help you do better experiments.

MileHigh